Development of viral surveillance for endemic disease, emerging viral characterization and zoonic spill-over spill-back models

• Main Author: Tan, Zi Hian
• Other Authors: Shu Jian Jun
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Science::Biological sciences::Genetics
• Online Access: https://hdl.handle.net/10356/161407
• Institution: Nanyang Technological University

With viral sequences being added into common online databases such as NCBI virus and GISAID, the genomic information is growing as sequencing tools advance and this enables more labs to obtain and share viral genomic information. However, sequence information is passive unless we know the patterns and traits of the viral genetic code and the implications on infected hosts and public health. This thesis presents the work done to extend the knowledge in using viral sequences to understand the behaviour of viruses. Viruses do not have to cause a global outbreak or have a high case-fatality ratio to be of concern. The dengue virus causes mostly subclinical infections; however, with as many as an estimated 300 million infections and 100million recorded dengue cases per year it brings about an economic and healthcare burden. As there is no vaccine available for the dengue virus, managing dengue infection relies solely on vector (mosquito) management. The first of three parts of this thesis presents a model for genomic surveillance of the dengue virus to predict its fitness which results in higher dengue cases. By analysing the codon usage patterns with historical clinical data, the dengue virus sequence collected during the month with high reported cases showed fitness adaptation to human and mosquito hosts. The fitness adaption in the dengue sequence can be applied to genomic surveillance of mosquitos for an early indicator of high dengue infection potential. Next, the second part of this thesis presents on the work to characterise a novel virus based on its sequence. In the past 2 decades, there have been 3 coronavirus outbreaks with the severe acute respiratory syndrome-related coronavirus 2 (SARS-CoV-2) causing a global pandemic. The sequences of SARS-CoV-2 are distinct from SARS-CoV and MERS-CoV, showing a closer relation to coronaviruses isolated in bats. A novel method of characterising viruses by studying the distribution pattern of amino acids encoded within 115,582 viral genomes. Viral genomes were determined to exhibit a Quanta distribution out of 4 functions. These distribution parameters are not random; the parameters exhibit a linear relation. Upon further analysis, the non-random behaviour extends to defining viral families or species. By analysing the sequences, the SARS-CoV-2 parameters lie on the natural linear line and groups near the SARS virus responsible for the outbreak in 2003. This result suggests that SARS-CoV-2 is a result of natural evolution. Next, by comparing the quanta distribution parameters of spike proteins, the model can predict that Pangolins have a possible role in the current COVID-19 outbreak. The third and final part of this thesis predicts the susceptible animals to SARSCoV- 2. Transmission from humans to animals cannot be overlooked as SARSCoV- 2 can rapidly spread within farmed and wild animals resulting in a new strain that can be transmitted back to humans, rendering vaccines ineffective. A group of animals was determined likely to be susceptible to SARS-CoV-2 and potentially act as an amplifying host. The model was compared with the similar closely related SARS virus and found that is it consistent with identifying the intermediate animal. Additionally, the result matches the recent mink farm outbreak where the stray cats could have a role in the co-infection with the minks. Our findings on susceptible animals aid in preventing human-to-animal transmission and provide insights into the animal origins of SARS-CoV-2.